# Real time trading analysis response with Machine Learning

Real time trading analysis response with Machine Learning

See below for the original email that started this response from one of my newsletter members:

–> I have never been a fan of these black box technologies

Others are

http://www.stratasearch.com/

http://www.priceactionlab.com/

Chaoshunter (chaoshunter.com)

Discipulus 5 (rmltech.com)

GeneXProTools (www.gepsoft.com)

I think the second point you asked about  is that yes I do use Scikit-learn and the various machine learning packages for data analysis including inference, modelling (prediction), and Bayesian approaches.

–> I believe SciKit is the smaller Machine Learning library compare to the more advanced ones like TensorFLow or Karas (I am also new to this ML stuff)

For machine learning you are often confronted with a large n

umber of features and the question as to which set of features are most significant. For this you use Principal Component Analysis for dimension reduction, (sometimes called Eigenanalysis) where you use a statistical procedure using an orthogonal transformation to convert a set of observations of possibly correlated variables into a set of values of linearly uncorrelated variables called Principal Components. The first principal component has the largest possible variance (that is, accounts for as much of the variability in the data as possible). This transformation is defined in such a way that the first principal component has the largest possible variance (that is, accounts for as much of the

variability in the data as possible), and each succeeding component in turn has the highest variance possible under the constraint that it is orthogonal (always at right angles) to the preceding components. The resulting vectors are an uncorrelated orthogonal basis set. PCA is used as an exploratory tool in data analysis and for making predictive models.

–> I keep hearing about these component and factors used

From Scikit-Learn we use the various machine learning algorithms, eg Support Vector Machine, Linear Discriminant analysis, Ada Boost etc…  via a set of randomized tests (test & train), and feed the results into a confusion matrix and  test the model prediction (actual vs. predicted). The constant consideration is that these machine learning tools are designed to deal with stationary data, whereas time series data is non-stationary. So stationarity of data is huge part of your analysis if you want good models.

Here is an example of an open source project where you can automate your Machine Learning in Python with TPOT and Genetic Algorithms.

See:

https://github.com/rhiever/tpot

http://rhiever.github.io/tpot/using/

https://github.com/rhiever/tpot/blob/master/tutorials/Titanic_Kaggle.ipynb

https://blog.alookanalytics.com/2017/05/25/automate-your-machine-learning/

–> Good to know all of the above!!!

# Real time trading analysis in Redis 4 with Lua Torch machine learning ?

I created this video in response

What about bonds? You briefly mentioned treasury bonds, but didn’t include them on your list. I’m particularly interested in first world countries that have an interest rate higher than many third world countries i.e Portugal. Moreover, I code in Java. It’s not as efficient as C++, but it’s quicker to code, because it’s a more conceptual language. The time saved in creating the code with Java is worth more than the efficiency gained with less memory used with optimized C++ code in most cases. Conditions in the market are constantly changing. An algorithm that worked yesterday might not work tomorrow. The slower coding of C++ may not be able to keep up with a changing market. In contrast, Java can allow one to more quickly adapt their strategy. What would be best is to create multiple programs for different types of markets. I like to think of them as different personalities. Perhaps what’s most important though, is an understanding of Bayesian math. That is, new data is constantly affecting probability. The Monty Hall problem is the most well known example of Bayesian math. Indeed, it’s Bayesian math more than anything which helps one understand why Java is better programming language than C++ in most cases.

A bunch of reminders including tonite:

Bryan Downing is inviting you to a scheduled Zoom meeting.

Time: May 21, 2017 7:00 PM Montreal

Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/519828492

Or iPhone one-tap (US Toll): +14086380968,519828492# or +16465588656,519828492#

Or Telephone:

Dial: +1 408 638 0968 (US Toll) or +1 646 558 8656 (US Toll)

Meeting ID: 519 828 492

International numbers available: https://zoom.us/

Just a reminder that the new Quant Analytics service is now fully underway with daily video trading calls based on my human workflow. The new benefit starting this Monday will be the private event on May 22 at 8PM EDT.

Bryan

P.S. Don’t forget as this is important to remember:

We are coming down to the wire over the next weeks where this  newsletter will be retired. To continue with the conversation, you will need to be doing the following:

Get my 2 ebooks to opt in for the automated newsletter found on my new server driven by Infusionsoft.

Go here

All future online events can be found here which points to my Facebook Page events page

NOTE I now post my TRADING ALERTS into my personal FACEBOOK ACCOUNT and TWITTER. Don't worry as I don't post stupid cat videos or what I eat!

# Response to Desperate person who wants a trading robot

Response to Desperate person who wants a trading robot

From an email response I got: Hi Bryan

I really don’t understand what you are talking about. Are you selling robot for trading forex or you teach your clients how to use robot to trade? I am asking these questions because I URGENTLY  need a trading assistant, that is a robot to help me trade successfully.
Kindly explain explain it to me properly.
Best regards
What more to say out of my people like this?

NOTE I now post my TRADING ALERTS into my personal FACEBOOK ACCOUNT and TWITTER. Don't worry as I don't post stupid cat videos or what I eat!

# Response to Tour of 18 screen MT4 forex trading system

Response to  Tour of 18 screen MT4 forex trading system

From an email I sent

A tour of 18 screen MT4 forex trading system

Response from a Quant Elite member

Here is the

I was curious about the multiple monitor statement.  To me, the beauty of automation is to reduce the need for multiple monitors.  A good software setup and platform can be the equivalent of dozens, or even hundreds of monitors, none of which need human interaction, unless some event is triggered.  So to me the point of automation is to not have to sit in front of a monitor wall all day.

Don’t forget  this people feel self important and trade like humans. If they have automation, they will just read some blog or forum, implement a coded up cloned strategy, and hope for the best. Gambling has better results I would think.

NOTE I now post my TRADING ALERTS into my personal FACEBOOK ACCOUNT and TWITTER. Don't worry as I don't post stupid cat videos or what I eat!

# Youtube video response on Linear Regression MapReduce in R and Hadoop

Thanks to Minstand on Youtube for this. It was a comment.     Example Script for a simple MapReduce Job with the RevolutionAnalytics rmr2-Package. Example and package source: github.com/RevolutionAnalytics/rmr2 Hadoop is running on a VirtualBox Image with Zentyal Server 3.01 (x86_64 GNU/Linux 3.2.0-29-generic#46Ubuntu). Installed Applications: • Java 1.6.0_27 (OpenJDK Runtime Environment, OpenJDK 64-Bit Server) • Apache Hadoop 1.0.4 ,Apache Pig 0.11.0, Apache Hive 0.9.0, Apache Thrift 0.9.0 • R 2.15.2 with rhbase 1.1, rhdfs 1.0.5 and rmr2 2.1.0 Installation and config made my hair grey, but finally I ran about 200jobs and everything’s fast&stable. 🙂 There was a video response on Youtube called Linear Regression MapReduce in R

NOTE I now post my TRADING ALERTS into my personal FACEBOOK ACCOUNT and TWITTER. Don't worry as I don't post stupid cat videos or what I eat!

# My response to an open source HFT trading platform in .NET

My response to an open source HFT trading platform in .NET

> Unfortunately being UK based, the time-zone difference makes these

> events very late at night (ie midnight).  Do you have any plans to

> record them?

>

> as a platform to research further, but I have a lot on my ToDo list,

> and already have SmartQuant so much of the functionality was already

> covered (though not Open Source).  I was about to mention another OS

> platform that I had done some investigation and seemed pretty good

> only to find that it has recently been purchased outright and now been

> re-branded by a vendor – shame.

>

> Keep up the good work!

Thanks for the kind words. I will be recording the webinars but only my membership will have access to it. It does cost money to stream those kind of long videos. I would like a better time for these webinars but these to be the same best for most people globally. ALso, XXX is the correct platform but ..so I am on the hunt for another platform but will try to experiment with a few but they will not be easily to integrate with.

Thanks

NOTE I now post my TRADING ALERTS into my personal FACEBOOK ACCOUNT and TWITTER. Don't worry as I don't post stupid cat videos or what I eat!

# My responses for potential Premium Members switching from Matlab to R, new strategies and algos, live webinars coming

My responses for potential Premium Members  switching from Matlab to R, new strategies and algos, live webinars coming.

-> I have spent a week on R and another on Linux Open source projects including NOSQL. I have seen the limitations platforms like Tradelink and Matlab bring to the table. I have broken through this which I will start rolling out my knowledge in it in coming weeks. Only Premium members will get this info. There is general info on the free parts of the site but good luck looking for all the components with over10000 postings. You also need to know what you are looking for. This info will be consolidated for Premium Members saving them many painful hours.

->            We all want strategies of course, but they are not so simple to develop as you know. They also take many months to develop but I can assure the value would be much higher than what is currently being paid for. Also, I switched from Matlab to R with over 55 research papers helping me to post these models/strategies to Premium Members only. It will save them many hours of development and implementation. I also switched to R because of Matlab’s expensive costs especially for remote algo development and clustering licensing. R is open source which means it is free. Also, I have recently discovered some components for Linux and C++ which can be used to build a very fast HFT platform. An example is

I hope this gives you an idea of how much research goes in before posting for Premium Members. Also, I am looking at a weekly online pow wow for my Premium members only. I want to get more live engagements for them so we can all learn what we are all doing but keep it in a private forum.